It is well known that complex electromechanical devices, such as computer disc drives, can be harmed by foreign substances which come into contact with vital components of the device. For example, dirt or dust particles which accumulate on the platters of a disc drive can damage the read/write head of the drive causing a "crash." Thus, such devices are typically manufactured within a clean room environment and are sealed prior to leaving the clean room to reduce or prevent the possibility of such contamination.
However, the current breed of disc drives spin much faster and are more densely packed with data than prior drives. These speed and size increases require that the read/write heads fly very close to the surface of the disc platters (on the order of a micron). In light of these very low fly heights, it is possible for matter smaller than common dust or smoke particles to cause head/disc crashes. Indeed, even chemicals or chemical compounds which are outgassed by the disc drive may be sufficiently large to interfere with the drive heads.
Although some disc drive components outgas chemicals and chemical compounds while the drive is inactive, the level of outgassing typically increases when the drive is operating and the components are exposed to high temperatures. These outgassed chemicals and chemical compounds are easily transported throughout the drive (due to the rotation of the disc platters and the resulting air currents within the sealed drive) where they typically bond to the substrate that coats the disc platters. In addition to physically interfering with the drive heads during operation of the drive, some outgassed compounds (e.g., adhesives) may react chemically with the drive heads during periods of inactivity when the heads are in direct contact with the disc platters. Such chemical reactions cause stiction between the heads and the disc platters which further contributes to early disc drive failure.
Thus, it is important for disc drive manufacturers (as well as manufacturers of other electromechanical devices which may be susceptible to damage from outgassed compounds) to carefully inspect all of the components which make up the drive for the presence of outgassed compounds. Examples of such components within a disc drive include motors, coil bobbins, magnets, adhesives and labels.
Inspections of such individual components are typically conducted by static headspace sampling where a component (such as a drive head) is placed within a small, sealed container and held at an elevated temperature until the outgassed compounds reach a state of equilibrium within the headspace. The term "headspace" is utilized herein to refer to the space within the sealed container which is not taken up by the tested component itself. The sealed container typically includes an open top sealed by a septum to allow a needle to penetrate the headspace and withdraw a sample of the equilibrated headspace. This sample is then analyzed using known techniques and equipment such as a gas chromatograph and a mass spectrometer to determine the composition of the different outgassed compounds.
However, this prior "static" approach suffers from a number of problems, foremost of which is that only a small amount of the headspace volume (approximately 1 milliliter) may be withdrawn by the syringe before the equilibrium within the sealed container is upset. This small sample reduces the sensitivity (i.e., increases the detection threshold) of the test so that the levels of the outgassed compounds may not be accurately measured, while other outgassed compounds may not be detected at all. A further drawback to the prior art static testing is that the sealed containers are typically of limited size so that larger components (such as disc drive spindle motors or coil bobbins) can not fit within the containers. These relatively large components are typically sectioned so that only a portion of the larger component is placed within the container. However, the cutting process, and the heat generated thereby, may contaminate the results of the headspace outgassing test. Furthermore, analyzing relatively small, exposed sections of larger components may artificially shield or increase important outgassing constituents.
The above problems associated with traditional "static" headspace outgassing tests have increased the interest in "dynamic" testing procedures. Simply put, a "dynamic" test utilizes a flow of gas within a testing container (i.e., within the "headspace" of the container) over a period of time to collect the outgassed compounds. This "carrier gas" is preferably an inert or neutral gas which does not react with any of the outgassed compounds. The inert gas thus carries the outgassed compounds from the headspace to the analytical equipment which analyzes the compounds. One example of a "dynamic" headspace sampling system is described in U.S. Pat. No. 5,646,334 entitled MULTISAMPLE DYNAMIC HEADSPACE SAMPLER, issued Jul. 8, 1997 to Scheppers et al., and assigned to the assignee of the present invention.
However, several aspects of the prior dynamic testing systems can be improved upon, including the accuracy and sensitivity of the test as well as the length of time required for the testing procedure. Specifically, prior dynamic testing containers typically comprise disposable jars such as glass mason jars having a threaded top. While the disposable glass jar is inexpensive, glass is not an inert material and thus the jar itself will contribute outgassed compounds over the course of the test, particularly as the jar is held at an elevated temperature for a number of hours. Next, an aluminum top having both an inlet and an outlet for the carrier gas is typically screwed to the top of the jar to define a sealed testing chamber. The upper location of both the inlet and the outlet reduces the "flushing efficiency" of the testing chamber since outgassed compounds at the bottom of the jar are not flushed from the testing chamber at the same rate as compounds at the top of the jar. Additionally, relatively heavy compounds at the bottom of the jar may not be captured at all due to the tendency of the carrier gas to remain in the upper portion of the jar. Furthermore, like the glass jar itself, the aluminum top is not inert and will also contribute to anomalous results. To account for the extra contributions from both the glass jar and the aluminum top, a "blank" must typically be included with each test run to determine the types and amounts of compounds outgassed by the container itself. The time required to test a blank container with each test run, together with the possible errors introduced with the analysis of each "blank," represents a large degree of inefficiency and uncertainty with the prior dynamic testing systems.
In addition to contributing their own outgassed compounds, the non-inert glass jar and aluminum top may also bond with those compounds which are outgassed by the sample, thereby reducing the sensitivity of the test. Furthermore, although the prior art containers typically include seals positioned between the jar and the top, the glass jar and the aluminum top have different expansion coefficients and thus tend to expand at different rates as they are heated. Such differing rates of expansion increase the likelihood of leaks which further contaminate the test results.
A further problem relating to "dynamic" testing systems is the requirement that the sample be maintained at an elevated temperature while still providing for gas lines running to and from the testing container. Previous dynamic testing systems address this problem by placing the containers upon a heated block during the duration of the test. However, while the heated block provides unimpeded access to the top portion of the container for connection of gas inlet and outlet lines, the block only applies heat to the bottom of the testing container which produces an undesirable temperature gradient within the sealed testing chamber (i.e., warmer at the bottom than at the top). The problem of uneven heating is further complicated by the position of both the inlet and the outlet for the inert carrier gas at the top of the container. Since the carrier gas itself is not heated, the inflow of the relatively cool gas at the top of the chamber further increases the temperature gradient between the top and bottom of the chamber. Uneven heating of the chamber makes it difficult to achieve equilibrium within the headspace and thus tends to dramatically increase the time required to collect a sample of the outgassed compounds, sometimes requiring up to 24 hours. Furthermore, the temperature gradient and the flow of a relatively cool gas at the top of the chamber tends to cause some of the outgassed compounds within the chamber to condense on the aluminum top, thereby further reducing the sensitivity of the test.
It is with respect to these and other background considerations, limitations and problems that the present invention has evolved.